Today, an increasing number of junior and senior high schools, and universities use computer-network-based distance learning systems. However, teachers face a cost problem when developing learning materials for such systems. It takes much time, practice, and devotion to design and develop learning materials from scratch. The cost is particularly high when teachers try to create distance learning systems to help students interactively and naturally. I believe that we can leverage knowledge bases and Semantic Web technologies for most of this work, and thereby largely relinquish it to computers. To justify this belief, we have developed a prototype system that can automate Web-based learning materials. The target problem is a mathematical optimization problem. This paper outlines the system and provides a concrete description of the automation process.